The Research On Algorithm Of Dynamic Weighing Electronic Analytical Balance

Electronic analytical balance is a kind of high precision measuring instruments which contains multidisciplinary technologies such as sensor technology, electromagnetism, analog and digital electronic technology, intelligent information processing, materials, structural mechanics, precision machinery and manufacturing. It is a quality standard of measurement which is widely used in national defense, pharmacy, quality control, and laboratory.Animal weighing is an important approach of research and it will seriously affect the results of the research topic. Moreover, the development of animal experiment will result in a high research level in many research field.Therefore, this thesis focuses on the algorithm research of dynamic weighing on the basis of 220g/0.1mg electronic analytical balance, and presents a fast fusion method of dynamic weight based on EMD and batch estimation. In order to simulate the accuracy of the algorithm,this thesis establishes a simulation model of dynamic weighing system. Finally combining with the laboratory of 220g/0.1mg electronic analytical balance verification, this thesis verifies the accuracy and practicality of the dynamic weighing algorithm,which provides a new way for electronic analytical balance to achieve rapid and accurate dynamic weighing.The main work of this thesis: firstly, this thesis introduces different classes of electronic analytical balances, and describes some dynamic weighing technology algorithms and their research value. Secondly, this thesis further introduces the overall structure of the 220g/0.1mg electronic balance system. According to the principle of system identification, this thesis estimates physical weighing data to transfer function of the correlation coefficient for sensors by using a batch method of least squares. Then, this thesis analyzes the causes of high and low frequency noise in dynamic weighing sampled signal,and systematically expound the basic concept of the digital filter and its application in dynamic weighing sampled signal denoising, and processes high and low frequency noise elimination of the actual dynamic weighing small animals sampled data with digital filtering method which based on Kaiser window filter and Kalman Filter. In order to weigh a living animal fast and accurately, a fast fusion method of dynamic weight based on EMD and batch estimation is presented in this paper. Firstly, the Empirical Mode Decomposition(EMD) methods is used to analyze the sampled signal of living animals dynamic weight, and the correlation coefficient of each order intrinsic mode function(IMF) and the sampled signal is calculated. Make the summation of the false IMF which distinguished by the correlation coefficient and the original residual amount as a new residual amount. After the batch estimation of the new residual amount, the accurate dynamic weight value can be obtained by eliminating the effects made by different kinds of noise and disturbance on the weighing results. In this paper, by means of establishing a simulation model of dynamic weighing system, simulation experiments are carried out upon the information fusion method of dynamic weighing based on combination of EMD and batch estimation. The results of the simulation experiments show that the algorithm has high accuracy. In order to verify the effectiveness of the fast fusion method of dynamic weighing based on EMD and batch estimation, a static weighing test of standard weights and dynamic weighing test in mice have been carried. The result showed that the method which proposed in this paper has high accuracy of which the weighing error is less than 0.5%.